Research map / Probabilistic and scientific modeling / Dynamical systems
Dynamic Compression Flows for Neuroscience Data
Ganchao Wei, Daniela F De Albuquerque, Miles Martinez, Shiyang Pan, John Pearson
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use DCF instead of a VAE for latent velocity decoding | DCF · Neural reach data (3D latent) | median R2: 0.242 -> 0.304 | Improved |
| use DCF instead of MARBLE | DCF · Neural reach data (3D latent) | median R2: 0.195 -> 0.304 | Improved |
| use DCF instead of CEBRA | DCF · Neural reach data (3D latent) | median R2: 0.202 -> 0.304 | Improved |
| use DCF instead of DMD | DCF · Neural reach data (3D latent) | median R2: 0.014 -> 0.304 | Improved |
| use DCF instead of T-PHATE | DCF · Neural reach data (3D latent) | median R2: -0.012 -> 0.304 | Improved |
| use DCF instead of LFADS | DCF · Neural reach data (3D latent) | median R2: 0.311 -> 0.304 | No clear effect |
| add nested dropout instead of soft shrinkage (ridge, LASSO, horseshoe) on latent scales (ablation) | DCF encoder · latent dimensionality control | effective dimensionality limit | Improved |
| vary loss weights (alpha, beta, eta) away from 1 (ablation) | DCF · Rotating-ball simulation | robustness of results | No clear effect |
Closest papers by meaning
- Probabilistic Decomposed Linear Dynamical Systems for Robust Discovery of Latent Neural Dynamics (NeurIPS 2024 · similarity 0.89)
- Modeling Latent Neural Dynamics with Gaussian Process Switching Linear Dynamical Systems (NeurIPS 2024 · similarity 0.89)
- Time-Evolving Dynamical System for Learning Latent Representations of Mouse Visual Neural Activity (NeurIPS 2025 · similarity 0.88)
- Learning Interpretable Hierarchical Dynamical Systems Models from Time Series Data (ICLR 2025 · similarity 0.88)
- Meta-Dynamical State Space Models for Integrative Neural Data Analysis (ICLR 2025 · similarity 0.88)
- Latent Diffusion for Neural Spiking Data (NeurIPS 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
